Cystic fibrosis (CF) affects 1 in 3,200 births and leads to debilitating lung disease and premature death at a median age of 32 years. Gene therapy may provide a cure for this disease by replacement of the defective protein, the cystic fibrosis transmembrane regulator (CFTR). The long-range objective of this application is the development of lung-targeted gene therapy for treatment of CF using viral vectors derived from a nonpathogenic human parvovirus, adeno-associated virus (AAV). AAV vectors can transduce airway and alveolar epithelial cells in the lungs of mice at rates of >20% and lasting for over 8 months. However, clinical trials involving AAV vector-mediated transfer of the CFTR gene to humans have yet to show useful levels of CFTR expression of clinical efficacy. This lack of expression is likely due to the difficulty of making AAV vectors that express CFTR because of the large size of CFTR cDNA, and the difficulty of measuring small changes in CF disease severity that might result from gene therapy. To circumvent these difficulties and to provide a more direct test of the utility of AAV vectors in humans, the specific aims of this proposal include clinical trials in CF patients that will address the safety, efficacy and immune responses to nasal and bronchial administration of AAV vectors that encode an easily detected histochemical marker gene, human placental alkaline phosphatase (AP). We have found that vectors derived from AAV serotype 6 show much improved transduction rates in mouse airway compared to commonly used AAV serotype 2 vectors, although more extensive safety data in humans is available for AAV2. Here we propose to compare expression of hpAP delivered by both AAV2 and AAV6 in CF subjects. Another specific aim will address the development of effective AAV vectors for transfer and efficient expression of the CFTR cDNA. These approaches are designed to most efficiently test and develop AAV vectors for treatment of CF.